Passive Indoor Visible Light Positioning (VLP) (Sep 2016 - Apr 2021)

Proposed a passive indoor visible light positioning (VLP) system that does not involve a user in the localization process i.e., the user does not hold a device or sensor tags. Derived theoretical bounds on the localization error and used deep learning framework in order to estimate the position of the user. Furthermore, proposed a fall detection system that detects the fall of a user in a room using a single set of impulse response measurements. The details pertaining to the proposed passive VLP systems can be found in the Publications section.

2D Direction-of-Arrival Estimation (Sep 2014 - May 2016)

Proposed 2D direction-of-arrival (DOA) estimation algorithms that use an L-shaped or parallel arrays in order to receive the signals. The proposed techniques use cross-correlation information of the received signals from non-coherent sources at the antenna arrays in order to estimate the elevation and azimuth angles of the source. The details pertaining to the 2D DOA estimation can be found in the Publications section.

Indoor Localization using WiFi (Jan 2012 - May 2014)

Proposed an indoor localization system that uses received signal strength (RSS) of WiFi signals in an indoor area. The proposed method requires a very small fraction of fingerprinting load (1% of total grid points, i.e. 2 out of 219 points in our case), some crowd sourced readings and plan coordinates of the indoor environment and employs machine learning in order to localize the user. Moreover, a few location estimates together with few fingerprints help to estimate the complete radio map of the indoor environment. The details pertaining to the WiFi-based indoor localization can be found in the Publications section.